Daily model included

daily.stan

@@ -0,0 +1,63 @@
+// The input data is a vector 'D' of length 'N' containing daily data. 
+functions {
+  real pc_phi_lpdf(real x, real lambda){
+    return -1.5*log(x) + log(lambda/2.0) - lambda/sqrt(x);
+  }
+}
+
+
+data {
+  int<lower=0> N_days;
+  int<lower=0> D[N_days];
+  int<lower=0> N_generation;
+  vector<lower=0>[N_generation] g;
+  real R0;
+  real I0;
+  real alpha;
+  real beta;
+  real lambda;
+}
+
+transformed data {
+}
+
+parameters {
+  vector<lower=0>[N_days] R_T;
+  real<lower=0> sigma_R;
+  real<lower=0> R_0;
+  real<lower = 0> phi;
+  real<lower=0> I_negT;
+}
+
+transformed parameters {
+  vector<lower=0>[N_days] mu;
+  real<lower=0.0> generation_value = 0.0;
+  mu[1] = D[1];
+  for(t in 1:N_days){
+      // Calculate the value of mu
+      generation_value = 0.0;
+      for(s in 1:N_generation){
+        if((t-s)>0){
+        generation_value+=g[s]*D[t-s];
+        }else{
+        generation_value+=g[s]*I_negT;
+      }
+      }
+     mu[t] = R_T[t]*generation_value;
+  }
+}
+
+model {
+  // Set the priors of interest. 
+  phi ~ pc_phi(lambda);
+  I_negT ~ normal(I0,sqrt(I0)) T[0,];
+  sigma_R ~ inv_gamma(alpha,beta);
+  R_0 ~ normal(R0,4.0) T[0,];
+  R_T[1] ~  normal(R_0,sigma_R) T[0,];
+  R_T[2:N_days] ~ normal(R_T[1:N_days-1],sigma_R);
+  // Data
+  D ~ neg_binomial_2(mu,phi);
+}
+
+generated quantities {
+}